Zinc-finger nucleases (ZFNs) are emerging as powerful tools for targeted gene modification in many organisms. The ability to make such changes in specific genes facilitates the analysis of gene function, the construction of experimental models of human disease, and ultimately the directed alteration of disease genes for human gene therapy. The current proposal addresses several questions that concern the best approach to these applications of ZFNs, using the fruit fly, Drosophila melanogaster, as the experimental organism. 1) A number of parameters will be tested for their effects on the efficiency of gene targeting, including the amount of homology required and the effects of including rather large insertions and deletions in the donor DNA. 2) Several aspects of the design of new zinc finger sets will be tested by application of the technology to several new Drosophila genes. 3) An extensive, deep-sequencing study will be done to determine where in the genome, in addition to the designed target, specific pairs of ZFNs make DNA breaks. These off-target cuts are the source of toxicity that has been detected for a number of ZFN constructs. Having a clear picture of where these sites are will allow derivation of rules regarding zinc-finger recognition in vivo and will facilitate detection of undesirable effects due to ZFN cleavage. 4) The ability of single-strand breaks to stimulate gene targeting will be tested. The double-strand breaks created by canonical ZFNs can have deleterious effects, as noted. Single-strand breaks may be safer, if they can stimulate targeting at reasonable levels without proceeding through double-strand-break intermediates. The results of all these studies will be directly relevant to ZFN targeting in other organisms, including human cells.